Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal TIN: TERAPAN INFORMATIKA NUSANTARA
Sigit Puspito Wigati Jarot
Sekolah Tinggi Teknologi Terpadu Nurul Fikri, Depok
Aerospace Engineering Arts Automotive Engineering Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Decision Sciences, Operations Research & Management Dentistry Earth & Planetary Sciences Economics, Econometrics & Finance Energy Engineering Environmental Science Health Professions Immunology & microbiology Industrial & Manufacturing Engineering Law, Crime, Criminology & Criminal Justice Materials Science & Nanotechnology Mathematics Mechanical Engineering Medicine & Pharmacology Neuroscience Nursing Physics Social Sciences Veterinary Other

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Evaluasi Montreal Forced Aligner dan Goodness of Pronunciation untuk Penilaian Pelafalan Bahasa Sunda Abdul Fatahillah; Sigit Puspito Wigati Jarot
TIN: Terapan Informatika Nusantara Vol 6 No 12 (2026): May 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/tin.v6i12.9869

Abstract

Sundanese is the second most widely spoken regional language in Indonesia, yet automated pronunciation assessment systems for this language remain extremely scarce. This study presents a systematic evaluation of the Montreal Forced Aligner (MFA) and Goodness of Pronunciation (GOP) pipeline for Sundanese pronunciation assessment within a prototype voice-based learning application. The dataset comprises 2,500 valid utterance samples collected from 50 native Sundanese speakers, covering 10 basa loma vocabulary items spanning 20 unique phonemes. MFA evaluation revealed total and systemic alignment failure: all 2,500 files (100%) were identified as problematic, with 17 of 20 phonemes consistently assigned exactly 10-millisecond durations. Three distinct parameter configurations produced identical failure rates (100%), confirming that the failures are intrinsic to MFA's limitations with very short-duration single-word audio (mean 0.69 seconds) for low-resource languages. GOP evaluation yielded a global top-1 accuracy of only 26.1%, characterized by anomalous dominance of the /l/ phoneme as top-1 for 14 of 20 phonemes. Functional testing demonstrated the system's inability to discriminate correct from incorrect utterances. On the technical side, the React Native and FastAPI prototype application was successfully implemented, with 6 of 8 black-box test scenarios passing. This research provides three principal contributions: (1) empirical contribution in the form of the first quantitative evidence that the standard MFA-GOP pipeline cannot be directly applied to Sundanese as a low-resource language with short-duration single-word audio; (2) methodological contribution in the form of an empirical baseline and replicable evaluation framework applicable to other regional languages of Indonesia; and (3) practical contribution in the form of a React Native–FastAPI client-server prototype that serves as a starting point for further development of Sundanese pronunciation assessment systems using alternative approaches.
Analisis Komparatif MLP dan GraphSAGE dalam Deteksi Bot Twitter/X pada Benchmark TwiBot-22 Mochammad Fikri Chaerul Chalik Ramdhan; Sigit Puspito Wigati Jarot
TIN: Terapan Informatika Nusantara Vol 6 No 12 (2026): May 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/tin.v6i12.9891

Abstract

Bot accounts on Twitter/X remain a significant challenge because they affect information integrity, distort public discourse, and complicate platform moderation. This article evaluates two bot detection approaches on the TwiBot-22 benchmark: a profile-feature-based Multilayer Perceptron (MLP) and a graph-based GraphSAGE model, using a 12-Stage Evaluation Framework that covers data validation, feature engineering, model training, threshold analysis, feature ablation, and multi-seed evaluation. The study is limited to an offline benchmark setting with 1,000,000 labeled accounts, 13.99% bots and 86.01% humans, and a fixed split of 70% training, 20% validation, and 10% testing. In the single-seed 15-feature comparison, MLP achieved F1(bot) of 0.53 and PR-AUC of 0.48, while GraphSAGE reached F1(bot) of 0.53 and PR-AUC of 0.46. In the confirmatory three-seed evaluation, the user_only_8 configuration produced F1(bot) of 0.53 and PR-AUC of 0.49 with lower variance, whereas all_15 produced F1(bot) of 0.53 and PR-AUC of 0.47 with higher variance. These findings indicate that the more economical profile-only configuration preserves practically identical binary-decision quality, offers better probability ranking quality, and shows lower variance. The main contribution of this article is a feature-economy argument: on TwiBot-22, added graph and feature complexity does not automatically yield proportionate practical gains.
Co-Authors Abdul Fatahillah Mochammad Fikri Chaerul Chalik Ramdhan